Overuse of pre- and post-surgery antibiotics in low-income countries may contribute to higher levels of resistant surgical site infections following gastrointestinal surgery
Globally, approximately 12% of patients develop a surgical site infection within 30 days of gastrointestinal surgery, according to a prospective cohort study of more than 12500 people in 66 countries, published in The Lancet Infectious Diseases journal.
The incidence of surgical site infection varied between countries depending on their development level, with patients in high-income countries being least at risk, and patients in low-income countries being most at risk.
The results also suggest that globally more than one in five (22%) surgical site infections were resistant to antibiotics given before surgery to prevent infections.
“These findings begin to characterise the relationship between surgical site infections and global antimicrobial resistance,” says Dr Ewen Harrison, NIHR Unit on Global Surgery at the University of Edinburgh, UK. “Worldwide, large amounts of antibiotics were consumed to prevent and treat surgical site infections, yet in a fifth of cases the causative microorganism was resistant to the pre-surgery antibiotics given, and this increased to one of three cases in low-income countries. This high prevalence illustrates a potentially important area for improvement worldwide, and reducing surgical site infections will help to ensure safe and essential surgery around the world.” 
The study tracked 12539 patients from 343 hospitals in 66 countries  who were undergoing elective or emergency gastrointestinal surgery to see whether they developed a surgical site infection within 30 days.
Overall, 59% of patients (7339 people) were from 30 high-income countries, 31% (3918) of patients were from 18 middle-income countries, and 10% (1282) of patients were from 18 low-income countries. 1.9% (235/12539) of all surgery patients died within 30 days of their operation, with the highest incidence in low-income countries (4.8%, 61/1282 patients died).
The most common types of surgery were the removal of the gall bladder or appendix, and half of the patients (49%) had emergency surgery. Of all patients in the study, 12% (1538/12539) developed a surgical site infection within 30 days of surgery. However, incidence varied depending on a country’s income level – with 9% (691/7339) of patients in high-income countries, 14% (549/3918) of patients in middle-income countries, and 23% (298/1282) patients in low-income countries developing surgical site infections.
The pattern remained even when taking into account the different patient characteristics, diseases, contamination levels, procedures, and hospitals in low-income countries.
Patients with a surgical site infection were more likely to die than patients without an infection (1.5% [162/11001] of patients with no infection died, compared with 4.7% [73/1538] of patients with a surgical site infection), and were more likely to have another infection and further surgery. In addition, hospital stays for patients with surgical site infections were three times longer on average than people without infections (7 days vs 2 days).
The authors also analysed how common antibiotic-resistant infections were. Microbiology results were available for 610 patients with a surgical site infection, and 22% (132 people) of cases were resistant to the antibiotics given before surgery to prevent infections. Again, incidence varied depending on a country’s income level – with patients in low-income countries most at risk of antibiotic-resistant surgical site infections (36%, 46/128), and high income countries least at risk (17%, 49/295).
Looking into the causes of this difference, the authors found signs of overuse of antibiotics in low-income countries. Patients in low-income countries were more likely to receive antibiotics before and after surgery than patients in middle- and high-income countries (before surgery antibiotic use occurred in 96% of patients in low-income countries, 87% of patients in middle-income countries, and 88% of patients in high-income countries, after surgery these figures were 86%, 80% and 46%, respectively). This trend remained even when the higher levels of surgery contamination in low-income countries were controlled for.
The authors note some limitations, including that it was not possible to follow up all patients to 30 days after surgery, particularly in resource-limited settings. It is therefore possible that some cases were missed; however, they note that the size of the study limits the likelihood of bias.